LGPRMLNov 28, 2023

Sinkhorn Flow: A Continuous-Time Framework for Understanding and Generalizing the Sinkhorn Algorithm

arXiv:2311.16706v12 citationsh-index: 7
Originality Incremental advance
AI Analysis

This work provides a theoretical foundation for improving optimization methods in machine learning, though it appears incremental as it builds on existing mirror descent formulations.

The paper tackles the problem of understanding and generalizing the Sinkhorn algorithm for entropy-regularized optimal transport by introducing a continuous-time framework, resulting in novel robust variants and a unified perspective on related dynamics.

Many problems in machine learning can be formulated as solving entropy-regularized optimal transport on the space of probability measures. The canonical approach involves the Sinkhorn iterates, renowned for their rich mathematical properties. Recently, the Sinkhorn algorithm has been recast within the mirror descent framework, thus benefiting from classical optimization theory insights. Here, we build upon this result by introducing a continuous-time analogue of the Sinkhorn algorithm. This perspective allows us to derive novel variants of Sinkhorn schemes that are robust to noise and bias. Moreover, our continuous-time dynamics not only generalize but also offer a unified perspective on several recently discovered dynamics in machine learning and mathematics, such as the "Wasserstein mirror flow" of (Deb et al. 2023) or the "mean-field Schrödinger equation" of (Claisse et al. 2023).

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes